modelscope/AgentEvolver
AgentEvolver: Towards Efficient Self-Evolving Agent System
AgentEvolver helps developers create and refine AI agents that can learn and improve their performance autonomously in various digital environments, like games or application interfaces. It takes an agent's initial capabilities and environment interaction data, then outputs a more skilled and efficient agent. This is designed for AI researchers and engineers who are building intelligent agents and want them to evolve their abilities continuously.
1,284 stars. Actively maintained with 2 commits in the last 30 days.
Use this if you are developing AI agents and need a framework to enable them to automatically generate tasks, learn from experience, and optimize their decision-making processes.
Not ideal if you are a non-technical user looking for a ready-to-use AI application, as this is a framework for building and training agents.
Stars
1,284
Forks
147
Language
Python
License
Apache-2.0
Category
Last pushed
Jan 30, 2026
Commits (30d)
2
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/agents/modelscope/AgentEvolver"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related agents
inter-co/science-codeevolve
CodeEvolve is an open-source evolutionary coding agent for algorithm discovery and optimization.
thinkwee/AgentsMeetRL
Awesome List for Agentic RL
e2b-dev/awesome-ai-agents
A list of AI autonomous agents
RenqiChen/Virtual-Scientists
[ACL 2025] Multi-Agent System for Science of Science
Human-Agent-Society/CORAL
multi-agent evolution organization for autoresearch and more